Administrators of computing infrastructure are often concerned about the effects of user-provided code on system performance and security. It is difficult to write code that does not contain defects. Defects in user-provided code may result in an excessive use of computing resources and/or an unauthorized use of computing resources. Furthermore, user-provided code may, purposely or inadvertently, include malicious code designed to harm a computing system or otherwise use a computing system in an unauthorized way. Consequently, administrators of computing infrastructure may require that user-provided code undergo a comprehensive security evaluation before it can be executed.